Expression of Adh8 mRNA is developmentally regulated in Japanese medaka (Oryzias latipes)

Comparative Biochemistry and Physiology, Part B 140 (2005) 657 – 664 www.elsevier.com/locate/cbpb

Expression of Adh8 mRNA is developmentally regulated in Japanese medaka (Oryzias latipes) Asok K. DasmahapatraT, Xueqing Wang, Mary L. Haasch Environmental Toxicology Research Program, National Center for Natural Product Research, Research Institute of Pharmaceutical Sciences, Department of Pharmacology, School of Pharmacy, University of Mississippi, 313 Faser Hall, P.O. Box 1848, University, MS 38677, USA Received 25 October 2004; received in revised form 6 January 2005; accepted 12 January 2005

Abstract We cloned two full-length alcohol dehydrogenase (ADH) cDNAs from the liver tissue of adult Japanese medaka (Oryzias latipes). The coding regions spanned 1134 and 1137 nucleotides (nt) and the deduced amino acid sequences shared 63.6% identity between them. Phylogenetic analysis of the deduced amino acid sequence data identified the 1137nt as an orthologue of mammalian Adh5 (Class III) and the 1134 nt as an ortholog of zebrafish Adh8 genes. Reverse transcriptase polymerase chain reaction (RT-PCR) analysis further showed that adult medaka Adh5 mRNA was expressed in all the organs tested (brain, eye, gill, GI, heart, liver, kidney, muscle, skin, spleen, testis and ovary) while Adh8 mRNA showed tissue-specific expression (eye, GI, liver, kidney, muscle and skin). Comparison of the Adh5 and Adh8 mRNA expression in eye, gill, liver, kidney and skin indicate that Adh8 mRNA copy numbers are higher in all these tissues compared to Adh5 mRNA expression. Both Adh5 and Adh8 mRNAs are expressed during embryonic development with Adh5 mRNA transcripts present with very high copy number throughout the development. However, Adh8 mRNA is expressed in very low copy numbers initially (~1 h post fertilization; hpf) but begin to increase from 48 hpf to a level of ~200-fold higher at hatching. Therefore, it appears that in Japanese medaka, the expression of Adh8 mRNA, not Adh5 mRNA, is developmentally regulated. D 2005 Elsevier Inc. All rights reserved. Keywords: Alcohol dehydrogenase; Brain; Cloning; Development; Gene expression; Japanese medaka; Liver; Kidney; Real-time PCR

1. Introduction Alcohol dehydrogenases (ADH; EC 1.1.1.1) are zinc containing dimeric cytosolic proteins belonging to the protein superfamily of medium-chain dehydrogenases/ reductases and consisting of a complex enzyme family with different forms and extensive multiplicity. ADHs are one of the first mammalian enzymes to have been purified and crystallized (Bonnichsen and Wassen, 1948; Galter et al., 2003). Enzymatic studies have led to the identification of different ADH isozymes, which are distinguished by substrate specificity and inhibitor resistance. After the identification of the corresponding genomic sequences,

T Corresponding author. Tel.: +1 662 9157077; fax: +1 662 9151285. E-mail address: [email protected] (A.K. Dasmahapatra). 1096-4959/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.cbpc.2005.01.007

isozymes are now grouped according to sequence similarity. In humans, seven different genes encoding ADH are known; they are all located in a single cluster on chromosome 4q21-25 (Yoshida et al., 1991). These seven genes have been grouped in five different classes and ortholog genes in other animals have been identified and characterized (Duester et al., 1999). Amino acid sequence analysis from multiple vertebrate species indicate that all ADH classes have evolved from one common ancestor presumably class III ADH, the only ADH found in lower animals, yeast and plants (Danielsson and Jo¨rnvall, 1992). ADH enzyme activity and nucleotide sequences have been reported from several fish species (Danielsson and Jo¨rnvall, 1992; Funkenstein and Jakowlew, 1996; Dasmahapatra et al., 2001; Reimers et al., 2004b). In zebrafish, and in gilthead seabream (Sparus aurata), Adh mRNAs are expressed during development (Funkenstein and

658

A.K. Dasmahapatra et al. / Comparative Biochemistry and Physiology, Part B 140 (2005) 657–664

Jakowlew, 1996; Dasmahapatra et al., 2001; Reimers et al., 2004b). While evaluating the toxic responses of ethanol in developing embryos of Japanese medaka (Oryzias latipes), we have observed that medaka embryos exposed to ethanol during development exhibit similar phenotypic features that are equivalent to the features observed in human babies with fetal alcohol syndrome (FAS). Therefore a fish developmental model, due to the external development and lack of maternal influence during development, might be a useful model to study FAS compared to current mammalian models. Recently, ethanol toxicity has been demonstrated in developing zebrafish embryo (Bilotta et al., 2004; Carvan et al., 2004; Loucks and Carvan, 2004; Reimers et al., 2004a) but the exact molecular mechanism of ethanol teratogenicity is unclear. Two theories have been proposed (Reynolds and Brien, 1995; Reimers et al., 2004b). Ethanol induces teratogenesis directly or indirectly through acetaldehyde formation or oxidative stress. To verify these hypotheses, as a first step, we decided to characterize alcohol-metabolizing enzymes in Japanese medaka. Alcohol dehydrogenase (ADH) enzyme activities have been reported during medaka development (Frankel, 1987) and several medaka partial cDNA sequences from EST analysis are reported in GenBank. The goal of this study was to clone ADH(s) and characterize Adh mRNA expression in Japanese medaka. Our results indicate that medaka has more than one class of ADH similar to cod and zebrafish and these ADHs show tissue-specific differential expression in the adult. Moreover, the ADHs are also expressed during development and Adh8 expression is developmentally regulated.

Total RNA, from embryos (10 embryos/sample) or tissues (20–500 mg) of adult medaka (both male and female) was prepared by using Trizol reagent (Invitrogen, Carlsbad, CA, USA), following manufacturer’s instruction. To each sample 0.5 mL of Trizol was added, tissue homogenized and centrifuged at 12,000g for 20 min at 4 8C. The upper clear phase was transferred to a fresh 1.5 mL centrifuge tube and 0.25 mL of isopropanol (Sigma-Aldrich, St. Louis, MO, USA) was added. After 10 min of incubation at room temperature, the mixture was centrifuged at 12,000g for 20 min at 4 8C. The precipitate was washed with 1 mL 75% ethanol. The washed RNA was made free of genomic DNA by treating the samples with DNase I (Promega, Madison, WI). One unit of DNase I was added to each RNA sample in a final volume of 50 AL of 1 transcription buffer (50 mM Tris–HCl, 75 mM KCl, 3 mM MgCl2, 0.5 mM MnCl2) and 40 U of RNasin (Promega, Madison, WI). The mixture was incubated at 37 8C for 20 min. After incubation, the DNase I was removed by the Trizol precipitation method as described before. RNA was stored at 80 8C. The concentration of RNA was determined in an Eppendorf Biophotometer and the purity of the RNA was checked by 1% agarose gel electrophoresis with 0.1% ethidium bromide. The samples with distinct 18S and 28S ribosomal RNA and A260/A280 ratio above 1.5 were used for analysis.

2. Materials and methods

2.3. Reverse transcriptase polymerase chain reaction (RT-PCR)

2.1. Experimental procedure Adult male and female medaka (3–4 months old and actively breeding) were maintained at 25 8C in balanced salt solution (BSS, 17 mM NaCl, 0.4 mM KCl, 0.3 mM MgSO4, 0.3 mM CaCl2) with standard diet and 16L: 8D photoperiod in the aquaculture facility located in the Pharmacology Department, Environmental Toxicology Research Program, School of Pharmacy, University of Mississippi. Newly fertilized embryos were collected in the morning (9:00 h) of the experimental day and maintained at 1 egg/mL hatching solution (17 mM NaCl, 0.4 mM KCl, 0.36 mM CaCl2 and 0.6 mM MgSO4) at 25 8C with 16L: 8D photoperiod in 48 well culture plates. Only viable and fertilized eggs were used in the experiment. The embryos were maintained for up to 12 days with a 50% static renewal of the media and the removal of the dead embryos every 24 h. Under these conditions the embryos generally began hatching at 7 days post fertilization (dpf). Any embryos not yet hatched after 12 days in culture were excluded from the experiments.

The Institutional Animal Care and Use Committee (IACUC) of the University of Mississippi (UM) approved all of the experimental protocols. 2.2. Isolation of RNA

The Quick Access RT-PCR kit (Promega, Madison, WI) was used for PCR amplification by following the manufacturer’s recommendations. Primers (Invitrogen, Carlsbad, CA) for the target genes of interest (Adh5 sense, 5V-CTTTTCGCCCTGAAGGAACC and antisense, 5V-GAGTTCAGGGCCACATATGATG-3V; Adh8 sense, 5VCATGGCCACAGCTGGTAAGG-3V and antisense, 5VGTTACGACTGAAGCGATATTG-3V) at 50 pM concentrations were added in a 25 AL reaction volume containing dNTPs, Mg2+, Tfl DNA polymerase and 100–200 ng of RNA. 2.5 U of AMV reverse transcriptase (Promega, Madison, WI) was added before the start of the reaction. Initially, two partial cDNA sequences from EST clones reported in GenBank (GenBank Accession AV668915 and AU179808) were used for designing primers. The reaction conditions were 48 8C for 45 min, 1 cycle; 94 8C 2 min, 1 cycle followed by denaturation at 94 8C for 30 s, annealing at 60 8C for 1 min and extension at 68 8C for 2 min, 40 cycles were done with a final extension at 68 8C for 7 min. The PCR products were purified by agarose gel (1%) electrophoresis and used for cloning and sequencing.

A.K. Dasmahapatra et al. / Comparative Biochemistry and Physiology, Part B 140 (2005) 657–664

2.4. 5V and 3V Rapid amplification of cDNA ends (5V and 3V RACE) 5V and 3V RACE techniques were used to obtain the start and stop codon, respectively. 5V and 3V RACE kits were purchased from Invitrogen and the cDNA was amplified following the manufacturer’s instruction. The first strand cDNA was synthesized from the medaka RNA with the primers provided in the kit (3V RACE) or designed (5V RACE). The cDNA for 5V RACE was purified, tailed with TdT in 1 buffer and 2 Al of the reaction product amplified with gene specific reverse primer and Abridged Anchor Primer provided by the supplier. The largest fragment of the product was cloned and sequenced to verify the presence of the start codon of the target gene of interest. For 3V RACE, 2 Al of prepared cDNA was amplified with a gene-specific forward primer of our design and a kit provided adapter primer. The largest fragment was cloned and sequenced to verify the presence of the stop codon of the gene. After obtaining the full sequence of the open reading frame, a new set of primers was designed for amplification of the entire coding region of both medaka Adh genes. 2.5. Cloning and sequencing The entire coding regions of the two Adh genes were cloned. We used the pGEM-T Easy Vector system (Promega, Madison, WI). Approximately 50 ng of the gel-purified PCR products was ligated into pGEM-T Easy vector by T4 DNA ligase and then transformed into JM109 Escherichia coli competent cells. The cells were plated on LB medium containing ampicillin. The plates were incubated at 37 8C overnight and white colonies were selected for analysis. The positive colonies were grown overnight in 3 mL LB media with ampicillin (Sigma-Aldrich) at 37 8C and the plasmid DNA was prepared by using a Miniprep DNA purification system (Promega, Madison, WI, USA). The nucleotide sequences of the plasmids or PCR products were analyzed by using a CEQ Dye terminator cycle sequencing quick start kit (Beckman Coulter, Fullerton, CA, USA) in a BeckmanCoulter CEQ 8000 Genetic Analysis System (Beckman Coulter). 2.6. Quantitative analysis of Adh mRNA For quantification of Adh mRNAs, we used quantitative Real-Time PCR techniques (qRTPCR). We performed realtime analysis in two steps. First, the target gene of interest was amplified by using 50–500 ng of total RNA in a 25 AL reaction volume using an Access RT-PCR kit (Promega) with target gene-specific forward (for Adh5 5V-GTCACACAGATGCCTACACTC-3V and for Adh8 5V-CATTGCTGGACGGACCTGGAAG-3V) and reverse primers (for Adh5 5V-GCCCCGGCAACTTTGCAGCCC-3V and for Adh8 5VGTCGGGAAACACTCAGGACTG), following manufacturer’s recommendation. The reaction mixture contained 5

659

AL of 5 buffer, 0.5 AL of 10 mM dNTPs, 1 AL of 25 mM MgSO4, 2.5 U of AMV reverse transcriptase, 2.5 U of Tfl DNA polymerase, 50 pM of forward and 50 pM of reverse primers, required amount of total RNA with a final reaction volume of 25 AL in 0.2 mL thin-walled PCR tubes (Midwest Scientific, St. Louis, MO). The reactions were performed on a PTC-200 thermal cycler (MJ Research, Reno, NV) at 48 8C for 45 min, one cycle, followed by heating at 94 8C for 2 min, one cycle and then denaturation at 94 8C for 30 s, annealing at 60 8C for 1 min and extension at 68 8C for 2 min for 12 cycles. The PCR products were stored at 20 8C. The amplified cDNA was diluted with nuclease free water (1: 20) and then 2.5 AL of the diluted sample was analyzed in real time PCR (Opticon 2, MJ Research, Reno, NV, USA) using DyNamok HS SYBR Green qPCR kit (Finnzymes, Espoo, Finland) following manufacturer’s instruction. The reaction mixture in 20 AL final volume contains 10 AL 2 buffer and 0.4 AL 50 buffer (provided in the kit) and 50 pM forward and 50 pM reverse primers of the target gene of interest and 2.5 AL of the diluted cDNA. The volume was adjusted with nuclease free water. The standards were prepared by RT-PCR as cDNA, using the same forward and reverse primers used in real-time PCR and the liver RNA, free of genomic DNA, prepared from adult fish, was used as template. The amplified PCR product was purified electrophoretically in 1% agarose gel containing ethidium bromide (0.1%), eluted and quantified in an Eppendorf Biophotometer. The standards were aliquoted in separate tubes, and stored at 20 8C until use. During real-time PCR analysis, the standards were diluted to requisite concentration with nuclease free water and used for the preparation of the standard curve. The reaction conditions were initial denaturation at 95 8C for 15 min, one cycle, followed by 40 cycles of denaturation at 94 8C for 10 s, annealing at 52 8C for 20 s for Adh5 and 60 8C for Adh8, extension at 72 8C for 30 s, fluorescence data collection for 1 s. After the end of each cycle the samples were incubated again for 1 s at 78 8C and a second set of fluorescence data collection for 1 s was made to prevent error due to the formation of primer-dimmers, if any. A final extension of one cycle at 72 8C for 10 min was made. The melting curve was constructed by plotting fluorescence data (1 s) against temperature (65 8C to 95 8C with an interval of 0.2 8C). If checking the samples using a 1% agarose gel electrophoresis were necessary (if the melting curve is not consistent with the standards), reannealing was done for one cycle at 72 8C for 10 min. Each reaction was performed in duplicate. The cycle threshold or C(t) line was set manually for the standard curve of each of the genes using the Opticon monitor software (MJ research, San Francisco, CA). Generally, the C(t) line was set at the point (fluorescence 0.05–0.1) where the signals surpass background noise and began to increase. This threshold was applied to all wells for consistent analysis of individual samples and standards for the experiments. Standards were run each time in every set of real-time PCR analysis. The results were expressed as mRNA copy number/ng of total

660

A.K. Dasmahapatra et al. / Comparative Biochemistry and Physiology, Part B 140 (2005) 657–664

RNA initially used for cDNA preparation. Agarose gel electrophoresis and melting curve analysis was used to confirm specific gene product formation. 2.7. Statistical analysis The data (mRNA copy number) were analyzed using one-way ANOVA with Tukey’s multiple comparison test. The results are expressed as meanFSE with pb0.05 considered as significant.

3. Results 3.1. Characterization of the cDNA clones encoding medaka ADH As a result of RT-PCR and RACE, two full-length cDNAs from adult medaka liver were cloned and sequenced. A 1276 bp cDNA with an open reading frame (ORF) of 1137 nt encoding 379 amino acid protein and a 1341 bp cDNA with an ORF of 1134 nt encoding 378 amino acid protein were obtained (Fig. 1). Further analysis of the data with ALPS alignment (http//:medakautgenome.org) identified both Adhs organized in 9 exons with a coding region spanning over 4–6 kb. Both cDNAs encode proteins having a calculated molecular weight of approximately 40 kDa, with 63.6% deduced amino acid sequence identity. To identify these proteins further, the amino acid sequences of medaka ADHs were aligned with ADH proteins reported

from other fish and mammalian species (Fig. 2). Alignments suggested that the 1276 bp cDNA coded for a protein with amino acid sequence identity similar to class III ADH (ADH5) proteins (94.7% with gilthead seabream, 89.2% with zebrafish ADH5, 88.9% with Baltic cod H form and 81% with Baltic cod L form) and the 1341 bp cDNA coded for a protein with amino acid sequence identity with classical fish ADH (ADH8) proteins with mixed properties (80.2% with Baltic cod ADH1, 71.8% with zebrafish ADH8A and 68.0% with zebrafish Adh8B). Moreover, medaka ADH5 showed 81.0% sequence identity with human ADH5 (Class III). However, medaka ADH8 showed 61.7% amino acid identity with human Class III, ~55% with Class I, 51.3% with Class II, 53% with Class IV and 51.5% with Class V of human ADH protein. After this characterization, the sequences were submitted to GenBank (Adh5, GenBank Accession AY512892 and Adh8, GenBank Accession AY68277). 3.2. Expression of Adh5 and Adh8 mRNA in adult medaka and during development RT-PCR analysis of RNA prepared from brain, eye, GI, gill, heart, kidney, liver, muscle, ovary, testis, spleen and skin of adult medaka (male and female) indicated that Adh5 mRNA was expressed in all the tissues studied (Table 1). However, Adh8 showed tissue-specific expression in liver, kidney, muscle, eye, GI and skin. Quantitative analysis (qRTPCR) of Adh5 mRNA in liver, heart, kidney, brain, ovary, eye, gill, GI and skin showed the ovary expressed the

Fig. 1. Deduced amino acid sequence of Adh5 and Adh8 mRNA of Japanese medaka. The mRNA sequences are available from GenBank (Accession No. AY512892-Adh5 and AY682722-Adh8). Numbers on the right represent amino acid number, starting with the methionine residue of ADH5 and ADH8. The amino acids shown in bold letters are different from each other.

A.K. Dasmahapatra et al. / Comparative Biochemistry and Physiology, Part B 140 (2005) 657–664

661

branchiostama (0.1618) cod ADH8 (0.1003) medaka ADH8 (0.0944) zebrafishADH8A (0.0904) zebrafishADH8B (0.1309) h ADH1(Class I) (0.0336) hADH2 (Class I) (0.0253) hADH3 (Class I) (0.0281) mADH1 (0.0816) hADH7 (Class IV) (0.0516) mADH7 (0.0580) hADH6 (Class V) (0.2060) hADH4 (ClassII) (0.1185) mADH2 (0.1600) hADH5 (Class III) (0.0388) mADH5 (0.0361) cod ADH-l (0.1083) cod ADH-H (0.0622) gilthead seabream ADH (0.0226) medaka ADH5 (0.0226) zebrafishADH5 (0.0544)

Fig. 2. The phylogenetic tree of medaka ADH5 and ADH8 with other fish and human ADHs as reported in GenBank. The tree was constructed by the neighbor-joining method (Saitou and Nei, 1986) using vector NTI analysis software. The numbers in parentheses indicate the calculated distance values that are related to the degree of divergence between the sequences. The GenBank accession numbers for the sequences used are: AAF73255 (Branchiostoma), P79896 (gilthead seabream), P26325 (Baltic cod), P81600 (cod H), P81601 (cod L), NP031435 (mouse ADH1), NP036126 (mouse ADH2), NP031436 (mouse ADH5), NP033756 (mouse ADH7), P07327 (human ADH1), NP000659 (human ADH2), NP000660 (human ADH3), NP000661 (human ADH4), P11766 (human ADH5), P28332 (human ADH6), JC7759 (zebrafish ADH5), AAP97853 (zebrafish ADH8A), AAP75382 (zebrafish ADH8B).

highest copy number of Adh5 mRNA followed by liver, GI, gill, heart, brain, eye, kidney and skin. In case of Adh8, the expression was highest in eye followed by liver, skin, kidney and the GI. The tissues where both Adh5 and Adh8 expression were observed, Adh8 mRNA copy number was approximately 1.5 to 1500-fold higher than that of Adh5 (Fig. 3).

The expression of these two Adh mRNAs was also observed during embryonic development by RT-PCR. Both Adh5 and Adh8 mRNA were expressed during embryonic development. Adh5 mRNA was detected in the embryos as soon as the development started (~1 hpf) and remained unaltered throughout the development to hatching (Fig. 4A). Adh5

Table 1 Expression of Adh5 and Adh8 mRNA in different tissues of adult Japanese medaka Organs

Adh5

Brain Eye GI Gill Heart Kidney Liver Muscle Skin Gonad (testis/ovary)

+ + + + + + + + + +

Adh8 + +

+ + + +

RNA was prepared from the tissues and gene-specific primers for Adh5 (sense 5V-CTTTTCGCCCTGAAGGAACC-3V and antisense 5V-GAGTTCAGGGCCACATATGATG-3V) and Adh8 (sense 5V- CATGGCCACAGCTGGTAAGG3V and antisense 5V- GTTACGACTGAAGCGATATTG-3V) were used to amplify the target by RT-PCR. The experiments were repeated four times with four separate fish (2 males and 2 females).

Adh mRNA (copy number/ng RNA)

200000

Adh8

100000

0 Eye

GI

Kidney

Liver

Skin

Fig. 3. Expression of Adh5 and Adh8 mRNA in different organs of adult Japanese medaka. The expression was determined by extracting total RNA from the organs and reverse transcribed by gene-specific primers (Adh5: sense 5V-CTTTTCGCCCTGAAGGAACC-3V and antisense 5V-GAGTTCAGGGCCACATATGATG-3V; Adh8: sense 5V-CATGGCCACAGCTGGTAAGG-3V and antisense 5V-GTTACGACTGAAGCGATATTG-3V). The required amount of cDNA was further analyzed by qRTPCR. The results are meanFS.E. of 3–5 separate experiments.

662

15000

Adh5 mRNA (copy number/ng total RNA)

A

A.K. Dasmahapatra et al. / Comparative Biochemistry and Physiology, Part B 140 (2005) 657–664

10000

5000

0 0

8

24

48

72

96

144 hatch

Hours of Development 10000

Adh8 mRNA (copy number/ng total RNA)

B

#

#

7500

# 5000

# #

2500

0 0

8

24

48

72

96

144 hatch

Hours of Development Fig. 4. Developmental expression of ADH5 and ADH8 mRNA in the embryos of Japanese medaka. RNA from ten pooled embryos at a specific time of development was extracted and reverse transcribed using genespecific primers. The required amount of cDNA was further analyzed by qRTPCR. The results were statistically analyzed by one way ANOVA followed by Tukey’s multiple comparison test and expressed as meanFSE of 4–6 separate experiments. Pb0.05 was considered as significant. In case of Adh5 the results were not significantly different from each other. However, in Adh8, an increase was noticed after 24 h of development. Pound symbol (#) indicates the results are significantly different from 0, 8 and 24 h of development. (A) Adh5. (B) Adh8.

However, Adh8 mRNA levels were undetectable during or after fertilization and remained undetectable by RT-PCR until 48 hpf. Adh8 mRNA qRTPCR analysis indicated that the expression of Adh8 mRNA was developmentally regulated (Fig. 4B). Adh8 mRNA copy number increased almost 200-fold from the early stages of development (0, 8 and 24 hpf) to hatching (Fig. 4B).

4. Discussion The results indicate that the liver of Japanese medaka expressed at least two ADH enzymes, which shared 63.6% amino acid identity between them (Fig. 1). Alignment of the deduced amino acid sequences with ADHs reported from other fish and mammalian species, identified these enzymes as the product of Adh5 and Adh8 genes (Fig. 2). The expression of Adh mRNAs was tissue-specific. Adh5 was expressed ubiquitously in all the tissues investigated (Table

1). This expression pattern is similar (nearly identical) to a classical class III type of expression pattern observed in humans for ADH5 (Estonius et al., 1996). Adh8 expression was restricted to liver, kidney, GI, muscle, skin and eye (Fig. 3). Interestingly, Adh8, like human ADH1 (Edenberg, 2000), was undetectable/unexpressed in the brain and heart tissues of adult medaka. Therefore, Adh8 of Japanese medaka may have similar functions as ADH1 in mammals. The zinc dependent medium chain alcohol dehydrogenases are cytosolic, dimeric isozymes that catalyze the reversible oxidation of alcohols to aldehydes or ketones, acting on a wide range of substrates, from methanol to longchain alcohols and sterols. Mammals have multiple medium-chain ADH that arose by repeated gene duplication. As a group, these enzymes are widely distributed both phylogenetically and in different tissues. The ADH genes differ in their pattern of expression. Therefore, they represent an excellent system in which investigation can be made to study the evolution of tissue specificity and developmental regulation (Edenberg, 2000). In previous studies in fish only two classes of ADH proteins have been reported. Class III ADH, the ancestor of all other ADHs, is found from bacteria to human. The second class of ADH is a mixed type, structurally similar with class III but functionally similar with class I (Danielsson and Jo¨rnvall, 1992). Evolutionarily, fish are the first vertebrate class with documented expression of more then one class of ADH. In cod of Baltic origin (Gadus morhua), the ethanol-active ADH is structurally similar to class III but functionally similar to class I. The second class of cod ADHs (both H and L forms) are a typical class III enzymes, with low alcohol dehydrogenase activity and high specificity for glutathione/formaldehyde (Danielsson and Jo¨rnvall, 1992). In zebrafish (Danio rerio), three Adh mRNA transcripts were identified (Dasmahapatra et al., 2001; Reimers et al., 2004b). Adh5 is structurally identical with class III, however Adh8A and Adh8B were shown to have the highest structural identity with cod Adh1 (now designated as Adh8) based on amino acid sequences. Functionally, zebrafish Adh8A metabolized alcohol but was not inhibited by 4-methyl pyrazole (4-MP), the classical inhibitor of class I ADH. Adh8B did not metabolize ethanol (Reimers et al., 2004b). On the basis of these features zebrafish Adh8A and 8B were best described as sharing a common ancestry with mammalian class I, II, IV and V alcohol dehydrogenases (Reimers et al., 2004b). Therefore, in fish, the evolution of Adh genes involves two distinct branches. One branch includes Class III ADH (Adh5), and the other branch as in Adh8, includes the remaining classes of ADHs (class I, II, IV and V). Although a functional analysis of medaka ADH protein was not undertaken, the functions of these proteins were predicted from their amino acid sequences. NAD-dependent dehydrogenases are some of the largest and most studied families of proteins. Several three-dimensional (3D) structures have been determined from X-ray crystallography, and

A.K. Dasmahapatra et al. / Comparative Biochemistry and Physiology, Part B 140 (2005) 657–664

all the ADHs are strikingly similar in topology. ADH proteins are relatively large with polypeptide chains folded in such a manner that they form two distinct domains. One domain binds the NAD cofactor, whereas the second domain anchors the alcohol substrate adjacent to the catalytic functional groups. The active site for these two proteins has been designated as the region between the two binding domains. Horse ADH was used for the first X-ray crystallography study (Eklund et al., 1976). In fish, Baltic cod ADH structure was determined by X-ray crystallography (Ramaswami et al., 1996). Based on alignment of the primary sequences of these two medaka Adh mRNA with cod ADH8, the two zinc-binding domains are found to be highly conserved; noncatalytic zinc atoms interact with four cysteines at position 99, 102, 105 and 113 (position numbers in cod ADH8 are 98, 101, 104 and 112) and the catalytic zinc atoms with cysteines at position 47 (46 in cod Adh8) and 176 (175 in cod) and with histidine at 69 (68 in cod). These structural similarities are also observed in other fish species (Funkenstein and Jakowlew, 1996; Reimers et al., 2004b). Another important structural feature is the presence of the Rossmann fold domain, which is found in numerous dinucleotide binding proteins that utilize FAD, NAD, and NADP (Bottoms et al., 2002). The Rossman dinucleotide-binding domain typically consists of three conserved glycine residues with the sequence GXGXXG. Since ADHs bind NAD, it was crucial to document the conservation of the Rossmann fold domain in medaka ADHs. The sequence GLGAVG found in both medaka ADH5 and ADH8 at position 201–206 is the possible Rossmann fold domain. Given these findings and the percentage amino acid identity between ADHs reported from other fish and mammalian species, we suggest that Adh8 of medaka may be functionally identical with cod ADH8 and zebrafish ADH8A, all having affinity for ethanol. Adh5 of medaka appears similar to a functional class III ADH, those being glutathione-dependent formaldehyde dehydrogenases (Edenberg, 2000). Both of the ADH mRNAs were expressed during embryonic development (Fig. 4A and B). Expression of Adh5 remained unaltered throughout development, however Adh8 showed an increase of about 200-fold in mRNA copy number prior to hatching. Expression of all classes of ADH in several organs during fetal development has been reported in mammalian models (Edenberg, 2000). In zebrafish expression of Adh5 and Adh8A and in gilthead seabream expression of Adh (class III) mRNA showed increases after hatching, but zebrafish Adh8B expression remained unaltered throughout development (Funkenstein and Jakowlew, 1996; Dasmahapatra et al., 2001; Reimers et al., 2004b). The expression of mRNA during development is the result of cellular differentiation and specialization. Appearance of specific enzymes during embryogenesis reflects metabolic and biochemical changes associated with increasing developmental complexity. In medaka, both ovary and testis expressed Adh5 mRNA and therefore high

663

levels of Adh5 mRNA transcripts in embryos soon after fertilization probably derive from parental sources as an embryonic reserve in the yolk. Functionally, class III ADH (Adh5) is a glutathione-dependent formaldehyde dehydrogenase. At present it is not known what the function of Adh5 mRNA is during embryonic development. Because the development of medaka embryos is external, it is possible that Adh5 mRNA protects the embryos from the effects of toxic chemicals of the environment or generated by normal metabolism. The gonads (testis and ovary) of medaka did not express Adh8; however, both eye and liver expressed the highest amount of Adh8 mRNA compared to the other organs studied (Fig. 3). We observed that Adh8 mRNA expression increased during embryonic development starting to increase during and after 72 hpf (Fig. 4B). This phenomenon may be related to the development and functions of the embryonic eye and liver. The pigmentation in the eyes and the formation of liver in medaka embryos starts during or after 72 hpf (Iwamatsu, 2004). Therefore, the increase in Adh8 may be a signal of normal development of the eye and liver during embryogenesis. Ethanol as a developmental teratogen produces cardiovascular malformations in the medaka embryo, which can be compared with cardiovascular abnormalities in FAS babies. We observed that the timing of ethanol exposure and the developmental stage of the embryo play a significant role in the induction of this cardiovascular abnormality (unpublished observations). The first 48 h of development is the critical period for cardiovascular development in medaka. Although the exact mechanism of ethanol toxicity is unknown, we hypothesized that ethanol toxicity (particularly the cardiovascular abnormality) in medaka is directly related to the metabolism of ethanol by alcohol metabolizing enzymes and the generation of ethanol metabolites. The present experimental findings showed that Adh5 mRNA expressed earlier than Adh8. However, ADH5 enzyme (class III ADH, the product of Adh5) is a glutathionedependent formaldehyde dehydrogenase, which can oxidize ethanol at high concentrations (Edenberg, 2000; Hoog et al., 2001). In human, the classical ethanol-metabolizing enzyme, ADH 1, starts to express predominantly in fetal liver. So ethanol metabolism in the human embryo can possibly occur during early development. In fish, the classical mammalian alcohol-metabolizing enzyme (ADH1) is yet to be identified, however, ADH8 may be regarded as the key enzyme responsible for ethanol metabolism (Danielsson and Jo¨rnvall, 1992; Reimers et al., 2004b). In zebrafish, the Adh5 and Adh8B with low affinity to ethanol expressed early and Adh8A with high affinity to ethanol expressed later in the development (Reimers et al., 2004b). Moreover, acetaldehyde, the metabolic product of ethanol, was found to be more toxic than ethanol (24 h LC50 for ethanol is 380.5 mM; acetaldehyde 0.541 mM) in the zebrafish embryo and able to induce similar defects as caused by ethanol (Reimers et al., 2004a). Therefore, these data suggest that induction of

664

A.K. Dasmahapatra et al. / Comparative Biochemistry and Physiology, Part B 140 (2005) 657–664

ethanol toxicity in fish is due to ethanol metabolites rather then a direct effect of ethanol. Further studies are needed to verify the hypothesis. From the present study in medaka, it is anticipated that Adh5, due its early expression, may play a key role in ethanol metabolism during early development (0–48 h) and the ethanol-induced cardiovascular abnormalities may be the result of acetaldehyde formation by ADH5 rather than direct effect of ethanol.

Acknowledgments This study was supported partially by the Office of Research and Sponsored Programs (small grant), a Chancellor’s Partner Grant and the Environmental Toxicology Research Program of the University of Mississippi. We are thankful to Jim Weston, a graduate student of the Department of Biological Sciences and Amit Chaudhury, a graduate student of the Department of Pharmacology, University of Mississippi, for their help in cDNA sequencing.

References Bilotta, J., Barnett, J.A., Hancock, L., Saszik, S., 2004. Ethanol exposure alters zebrafish development: a novel model of fetal alcohol syndrome. Neurotoxicol. Teratol. 26, 737 – 743. Bonnichsen, R.K., Wassen, A.M., 1948. Crystalline alcohol dehydrogenase from horse liver. Arch. Biochem. Biophys. 18, 361 – 363. Bottoms, C.A., Smith, P.E., Tanner, J.J., 2002. A structurally conserved water molecule in Rossmann dinucleotide-binding domains. Protein Sci. 11, 2125 – 2137. Carvan III, M.J., Loucks, E., Weber, D.N., Williams, F.E., 2004. Ethanol effects on the developing zebrafish: neurobehavior and skeletal morphogenesis. Neurotoxicol. Teratol. 26, 757 – 768. Danielsson, O., Jfrnvall, H., 1992. Enzymogenesis: classical liver alcohol dehydrogenase origin from the glutathione-dependent formaldehyde dehydrogenase line. Proc. Natl. Acad. Sci. U. S. A. 89, 9247 – 9251. Dasmahapatra, A.K., Doucet, H.L., Bhattacharyya, C., Carvan III, M.J., 2001. Developmental expression of alcohol dehydrogenase (ADH3) in zebrafish (Danio rerio). Biochem. Biophys. Res. Commun. 286, 1082 – 1086. Duester, G., Farres, J., Felder, V.C., Hoog, J.O., Pares, X., Plapp, B., Yin, S.Y., Jfrnvall, H., 1999. Recommended nomenclature for

vertebrate alcohol dehydrogenase gene family. Biochem. Pharmacol. 58, 389 – 395. Edenberg, H.J., 2000. Regulation of the mammalian alcohol dehydrogenase genes. Prog. Nucleic Acid Res. 64, 295 – 341. Eklund, H., Nordstrom, B., Zeppezauer, E., Soderlund, G., Ohlsson, I., Boiwe, T., Soderberg, B.O., Tapia, O., Branden, C.I., Akeson, A., 1976. Three-dimensional structure of horse liver alcohol dehydrogenase at 2– 4 2 resolution. J. Mol. Biol. 102, 27 – 59. Estonius, M., Svensson, S., Hoog, J.O., 1996. Alcohol dehydrogenase in human tissues: localisation of transcripts coding for five classes of the enzyme. FEBS Lett. 397, 338 – 342. Frankel, J.S., 1987. Asynchronous expression of alleles at the alcohol dehydrogenase locus during Oryzias hybrid development. Zool. Sci. 4, 735 – 737. Funkenstein, B., Jakowlew, S.B., 1996. Molecular cloning of fish alcohol dehydrogenase cDNA. Gene 174, 159 – 164. Galter, D., Carmine, A., Buervenich, S., Duester, G., Olson, L., 2003. Distribution of class I, III and IV alcohol dehydrogenase mRNAs in the adult rat, mouse and human brain. Eur. J. Biochem. 270, 1316 – 1326. Hoog, J.O., Hedberg, J.J., Stromberg, P., Sevensson, S., 2001. Mammalian alcohol dehydrogenase—functional and structural implications. Biomed. Sci. 8, 71 – 76. Iwamatsu, T., 2004. Stages of normal development in the medaka Oryzias latipes. Mech. Dev. 121, 605 – 618. Loucks, E., Carvan III, M.J., 2004. Strain-dependent effects of developmental ethanol exposure in zebrafish. Neurotoxicol. Teratol. 26, 745 – 755. Ramaswami, S., El Ahmad, M., Danielsson, O., Jfrnvall, H., Eklund, H., 1996. Crystal structure of cod liver class I alcohol dehydrogenase: substrate pocket and structurally variable segments. Protein Sci. 5, 663 – 671. Reimers, M.J., Flocktom, A.R., Tanguay, R.L., 2004a. Ethanol- and acetaldehyde-mediated developmental toxicity in zebrafish. Neurotoxicol. Teratol. 26, 769 – 781. Reimers, M.J., Hahn, M.E., Tanguay, R.L., 2004b. Two zebrafish alcohol dehydrogenases share common ancestry with mammalian class I, II, IV, and V alcohol dehydrogenase genes but have distinct functional characteristics. J. Biol. Chem. 279, 38303 – 38312. Reynolds, J.D., Brien, J.F., 1995. Ethanol neurobehavioural teratogenesis and the role of l-glutamate in the fetal hippocampus. Can. J. Physiol. Pharm. 73, 1209 – 1223. Saitou, N., Nei, M., 1986. The number of nucleotides required to determine the branching order of three species, with special reference to the human-chimpanzee-gorilla divergence. J. Mol. Evol. 24, 189 – 204. Yoshida, A., Hsu, L.C., Yasunami, M., 1991. Genetics of human alcohol-metabolizing enzymes. Prog. Nucleic Acid Res. Mol. Biol. 40, 255 – 287.